Fusion: Practice and Applications

Glaucoma is a condition where the eyes of human beings are infected due to retinal damage which could result in loss of vision. It generally occurs due to prolonged pressure on the eye and affects the optic nerve if not treated at the earliest stage. However, it is hard for even experts to detect it at the earlier stage. Hence numerous image processing techniques were applied to identify Glaucoma in retinal eyes. The profound purpose of the work is to propose a pre-processing console to remove outliers in the Glaucoma retinal Fundus images using Denoising techniques of pre-processing to enhance the prediction using image pre-processing and computer vision techniques. The model was created with three stages including applying the denoising model using the Median Filtering for Edge Preservation, Contrast Limited Adaptive Histogram Equalization (CLAHE) and optimizing by eliminating irrelevant features using the Black Widow Optimization model and finally evaluating the performance of denoising techniques using accuracy-based predictions. The results showed that after performing a combination of denoising and optimizing techniques, the image quality was enhanced with 97% outperforming the existing models.  

In a cloud context, merging complimentary numerous virtual machines (VMs) on an existing physical machine (PM) is the primary method for optimizing physical resources. One well-known area of research concentrates on making better use of VM migration resources when taking into account the dynamically changing resource demands of VMs. Finding the ideal balance between the complexity and performance of the VM migration optimization is the problem here. On the one hand, effective resource reuse is achieved through VM migration planning, and on the other, VM migration frequency is decreased to improve migration efficiency. On the other hand, a cloud data centre’s enormous PM and VM population typically makes migration planning more challenging, which impedes the VM migration decision-making process. By reducing the number of VM migration options to make VM migration planning easier and address these issues, this study recommend a hybrid Ant Colony and Genetic Algorithm (AGO) resource pool architecture. Then, establishing this model as a base, we develop the hybrid resource-reuse optimization method, which maximizes resource utilization with a minimal number of VM migrations. Finally, we evaluate hybrid AGO using simulation testing and real-world trials on a working cloud platform. Compared to similar methods, the findings show that hybrid AGO increases average resource utilization by 15%, reduces the use of PMs by 15%, and decreases the average number of migrations by 30%.

In video games, artificial intelligence is the effort of going beyond scripted interactions, however complex into the arena of truly interactive systems. To make a game world appear more real, these video games must be responsive, adaptive, and intelligent. For example, in real time strategy games, if there is an enemy seeking/hunting the player, it will be moving in paths, turning around and even maybe jumping in order to find the player. In this case, if the enemy acts/moves more real like human, it will be a benefit for making the game more attractive and exciting. This paper aims to develop a fast, intelligent, and realistic pathfinding approach that makes a user feel that he/she is playing with a human being instead of a machine. To achieve this, this paper presents a Heap Heuristic A* Algorithm as an enhancement of A* algorithm, in which the Chebyshev distance is used to control the smoothness of the resulted path and heapsort algorithm to sort the nodes easily without a lot of memory consumption. Compared to the pervious improved A* algorithms, the proposed algorithm produces a smoother path while consuming less memory to get a final result of human like movement. The experiments results showed that the proposed algorithm reduced the computing time by 66.6% using a grid size of 200*200 compared with A*MOD algorithm. Also, they showed that the proposed work takes almost 91ms to find the path compared to 363 ms and 116 ms when Native A* and A*MOD algorithms are used, respectively, Furthermore, the proposed algorithm performance remains stable in the case of increasing the number of visited nodes, despite the changing order of obstacles.

The fusion of computer technologies has had a remarkable impact on contemporary culture, as computers have a substantial impact on practically all aspects of learning; nonetheless, some students have claimed that they still feel uncomfortable when using computers. Test anxiety related to computer–assisted assessment (CAA) is a main factor that is expected to influence students’ academic achievement. Learning math in the digital environment could be a challenging process for students which could increase anxiety levels among them. The current quantitative research study pursues to measure students’ levels of anxiety that result from learning and assessment with computers and discover whether anxiety level is associated with students' academic achievement in tertiary institutions. Descriptive analysis and Correlation Coefficient are the employed statistical techniques to achieve the study objectives. Findings demonstrated that more than 90% of the sample identified with low anxiety levels and there is a noteworthy negative correlation between anxiety levels and students’ academic achievement in math. The findings have implications for practice in the higher education sector in instructional design and university counselling services.

Internet-of-Things (IoT)-based heart disease prediction is a complex task and processing the real collected data directly for remote patient monitoring suffers from the limitations due to the irrelevant data features, affecting the prediction accuracy and raising the security concerns. Hence, the efficient Adaptive ensembled deep Convolution neural network –Bidirectional Long Short Term Memory (Adaptive ensembled deep CNN-BiLSTM ) classifier model is proposed via the fusion of interactive hunt-based CNN and Whale on Marine optimization (WoM)-based deep BiLSTM. The Adaptive optimization developed from the standard hybrid characteristics such as random searching, seeking, attack prohibition, following, and waiting characteristics optimized the fusion parameters of the developed classifier for attaining high detection accuracy. Additionally, the modified Elliptic Curve Cryptography (ECC) based Diffi-Huffman encryption algorithm provides the authentication and security of sensitive patient data in heart disease prediction. The developed model is evaluated with other competent methods in terms of accuracy, sensitivity, specificity as well as F-measure, which are reported as 97.573%, 98.012%, 97.592%, and 97.705% respectively.

Due to advancement in technology, various fields have boosted the development of systems that improve people’s life quality, contributing to the welfare of the community by providing relevant and pertinent information for decision-making. On the Internet of Things (IoT), the systems demand measuring and monitoring several environmental variables. The heterogeneity of the captured data and the measuring instruments used to hinder the interoperability among the different components of the IoT. The problems are raised an interest in the development of methods and tools that support the heterogeneity of the data from the sensors, the measurements, and the measuring devices. Some existing tools have resolved some of these interoperability problems.  However, it forces to IoT developers to use sensors from specific brands, limiting their generalized use in the community. Furthermore, it is required to solve the challenge of integrating different protocols in a same IoT project. Besides, by generating alerts, it may help making decisions daily, considering the data provided by the sensors. it is required to solve the challenge of integrating different protocols in a same IoT project. To overcome the limitations of the existing glitches, there is need to develop a framework based on network of sensors via software that allows communication-using protocols in a specific environment to monitor the quality of air and to alarm users about this. In this paper, a prototype of proposal is mentioned about the architecture, list of hardware, software and different APIs are utilized to gather data in a systematic way so as users can visualize data in a semantic view. The visualization is shown later by using Matplotlib, Seaborn tools of Machine Learning (ML) and Deep Learning (DL) to plot the temperature along with humidity in a historical span. The result shows that accuracy obtained via Machine Learning Classifier is 87% in the context of Weather Prediction.

Due to the very high direct or indirect costs of fraud, banks and financial institutions seek to accelerate the recognition of the activities of fraudsters. The reason for this is its direct effect on serving the customers of these institutions, reducing operating costs and remaining as a reliable and valid financial service provider. On the other hand, in recent years, with the development of information and communication technology, electronic banking has become very popular. In the meantime, it is inevitable to use fraud detection techniques to prevent fraudulent actions in banking systems, especially electronic banking systems. In this paper, a method has been developed that leads to the improvement of fraud detection in information security and cyber defense systems. The main purpose of fraud detection systems is to predict and detect false financial transactions and improve the intrusion detection system using information classification. In this regard, the genetic algorithm, which is known as one of the stochastic optimization methods, is used. At the end, the results of the genetic algorithm have been compared with the results of the decision tree classification and the regression tree. The simulation results show the effectiveness and superiority of the proposed method.  

Accurate forecasting of future electricity consumption is necessary to create a satisfactory design for an electricity distribution system. To enhance forecasting accuracy, autoregressive integrated moving average (ARIMA) was compared with hybrid of ensemble empirical mode decomposition (EEMD) plus autoregressive integrated moving average (ARIMA) denoted by (EEMD+ARIMA), to know which model is better performing a historical US monthly electricity consumption from DEC-2000 to SEP-2022 were used. The data were divided into training set (90%) and testing set (10%) to insure the model accuracy. The mean absolute square error, root mean square error, mean absolute error and mean absolute percentage error measurements were used to test the ARIMA and hybrid EEMD+ARIMA performance, the results show that the hybrid EEMD+ARIMA outperforms ARIMA model with the lowest RMSE, MAE, MPE, MAPE, MASE. For the best model, Akaike Information Criterion and Bayesian Information Criterion were applied to choose the best. The results show that the AIC and BIC of the EEMD+ARIMA were lower than the ARIMA model, which indicates that the EEMD+ARIMA is better than the single ARIMA in forecasting of electricity consumption. The conclusion reveals that the hybrid EEMD+ARIMA provides more accurate forecasting and performs significantly better than the ARIMA in forecasting of electricity.